Dynamic credit allocation for researchers

Abstract

Credit allocation of each author of a multiauthor paper has been a long standing concern. Regarding to the fact that the credit of each citing paper is not equal and the total credit of paper fades with time, in this paper, we introduce the PageRank algorithm and time parameter, and present an improved dynamic credit allocation method, namely IDCA method. By investigating the effect of the damping factor d of PageRank algorithm and the time parameter γ on the accuracy α of the IDCA method, finally, the two parameter (d,γ) are set (0.85,-3). Then, we validate the method by distinguishing the laureates of the Nobel Prize in Physics from the authors of prize-winning papers in American Physical Society(APS) dataset. Result shows that the IDCA method outperforms the state-of-the-art methods, and the accuracy of identifying the Nobel Prize laureates in Physics is 80.77% for 26 multiauthor prize-winning papers in APS dataset. Furthermore, we test on the real dataset with noise data generated from randomly adding and rewiring edges, and results indicate that the IDCA method still has higher accuracy compared with the state-of-the-art methods. Finally, by studying the temporal evolution of credit among coauthors in the Nobel-winning paper, we find that the effect of the Nobel Prize on credit share is remarkable. This method is not just valid for physics, also valid for any other branch of sciences.